autoantibodies in the autoimmune rheumatic diseases

What’s new?

C The role of anti-CCP antibodies in the diagnosis of RA has

become more firmly established

C There is increasing recognition that rising concentrations of

either anti-dsDNA antibodies or ANCA should not be the sole

ground for altering treatment

C The emerging role of anti-b2-glycoprotein antibodies in the

diagnosis of the antiphospholipid antibody syndrome

ASSESSING THE PATIENT

Autoantibodies in theautoimmunerheumatic diseasesRichard Watts

AbstractThe detection of autoantibodies to a variety of antigenic targets plays an

important role in the diagnosis of autoimmune rheumatic disease. Rheu-

matoid factor, antinuclear antibodies and antineutrophil cytoplasmic anti-

bodies have been the key in the diagnosis of rheumatoid arthritis (RA),

systemic lupus erythematosus and systemic vasculitis for many years.

Antibodies against cyclic citrullinated peptides have been shown to be

specific for rheumatoid arthritis and may predict the development of

erosive disease, and are being increasingly used in the assessment of

patients with RA. There is increasing interest in the use of anti-C1q anti-

bodies for the detection and monitoring of lupus nephritis as this assay

appears to be more sensitive than anti-DNA antibodies. Antibodies

against b2-glycoprotein are being used more frequently as part of the

diagnosis of the antiphospholipid antibody syndrome. Enzyme-linked

immunosorbent assays are now being used in place of both radioimmu-

noassay and immunofluorescence because they are readily automated

and less dependent on the use of experienced technicians.

Keywords ANA; ANCA; anti-CCP antibodies; autoantibodies; ELISA;

rheumatoid factor

The detection of autoantibodies in the serum of patients with

suspected autoimmune rheumatic disease is an important part of

the diagnostic process. The clinical utility of the results is

dependent on the quality of the laboratory test. Many assays are

now available commercially. Access to a reference laboratory is

essential in difficult cases, particularly when there is a discrep-

ancy between the clinical features and laboratory results.

An ideal diagnostic test has both high sensitivity and high

specificity; that is, it identifies all patients with the disease (high

sensitivity) and is not present in those who do not have the

disease (high specificity). A number of different methods are

used to detect autoantibodies with varying specificity and

sensitivity. The most widely used methods in routine practice are

indirect immunofluorescence (IIF) and enzyme-linked immuno-

sorbent assay (ELISA). ELISAs can be readily automated and are

therefore suited to use in a routine non-specialized laboratory. In

addition, they provide an element of quantification of the auto-

antibody response. In general, they are more sensitive and less

specific than other methods such as immunoprecipitation or

immuno-electrophoresis.

Richard Watts MA DM FRCP is Consultant Rheumatologist at Ipswich

Hospital NHS Trust and Clinical Senior Lecturer at the University of East

Anglia, UK. Competing interests: none declared.

MEDICINE 38:2 69

Rheumatoid factor

Rheumatoid factors (RF) are antibodies directed against the Fc

part of the patient’s own immunoglobulin G (IgG) molecules.1

Rheumatoid factors are traditionally detected using an aggluti-

nation assay with gelatin particles coated with denatured IgG,

but ELISAs are being increasingly used. Although 80% of

patients with rheumatoid arthritis are RF positive, these anti-

bodies are not specific for rheumatoid arthritis and occur in

a wide range of autoimmune rheumatic diseases and infections,

usually at low levels.

Anticyclic citrullinated peptide (CCP) antibodies

The relatively poor specificity of traditional RF assays for RA has

led to the development of new serological markers.2 The most

promising of these are antibodies directed against citrullinated

antigens,3 such as vimentin, fibrinogen and cyclic citrullinated

peptides. Their specificity is determined by their ability to bind

citrulline, a post-translationally modified arginine residue. Anti-

CCP antibodies as detected by ELISA are present in 80% of

patients with established RA. The specificity is 85e90%, with

sensitivity of 50e60%. Furthermore, a positive anti-CCP test

appears to predict the development of erosive RA and this

predictive value complements that of RF.2,3

Antinuclear antibodies

Antinuclear antibodies (ANA) are autoantibodies directed against

cellular proteins or nucleic acids.4e6 Despite the name, a number

of these antigens are partly or wholly confined to the cell cyto-

plasm. The most frequently occurring ANA react with DNAe

protein or RNAeprotein complexes. These autoantibodies, which

are usually high-titre, high-affinity IgG antibodies, are generated

by a T-cell-dependent process driven by the autoantigen.

Screening for ANA is an important initial stage in the assess-

ment of autoimmune rheumatic disease. The test is usually per-

formed by IIF on fresh frozen sections of rodent liver and/or

kidney or on cultured human cell lines, such as HEp-2. HEp-2 cells

are preferred because individual cells and their organelles are

easily observed. They divide rapidly and present antigens which

are only sparsely present or absent in the resting nuclei of tissue

sections. Amongst the RNAeprotein complex targets, anti-Ro

antibodies are directed against the human antigen and may not be

detected on rodent sections. Several immunofluorescence staining

patterns occur, reflecting different antigenic specificities

(Figure 1). The IIF ANA may be falsely negative if the antigen is

� 2009 Published by Elsevier Ltd.

Figure 1


located outside the cell nucleus (e.g. anti-Jo-1 and antiribosomal

P), or if the antigen is present in a form not recognized by the

autoantibody (e.g. anti-Ro directed against epitopes on the Ro

molecule not expressed in cultured HEp-2 cells). The assay

requires an experienced technician to read the slides and results

are prone to subjective interpretation. Therefore, in many non-

specialized laboratories ELISA assays are being used to detect

ANA. These assays vary in sensitivity and specificity both between

different ELISA assays and between ELISA and IIF. Semi-quanti-

tative evaluation can be obtained by serial dilutions to end-point

fluorescence. The titre is the reciprocal of the dilution. A titre of

>1:80 is suggestive of autoimmune rheumatic disease.

Anti-DNA antibodies

The detection of antibodies to double-stranded (ds) DNA is

central to the diagnosis of systemic lupus erythematosus (SLE).7

Anti-dsDNA antibodies rarely occur in other diseases. In

contrast, antibodies to single-stranded DNA are found in a wide

variety of autoimmune and infectious conditions. Anti-dsDNA

antibodies can be detected using several methods. The most

specific is IIF using the haemoflagellate Crithidiae lucillae, which

has a kinetoplast containing circular dsDNA, and this is consid-

ered the ‘gold standard’. The ELISA is rapid, sensitive and

quantifiable, but less specific than the Crithidia assay. It is crit-

ically dependent on the use of pure dsDNA.

Antibodies to extractable nuclear antigens

Detection of antibodies against the extractable nuclear antigens

(ENA) is useful in the diagnosis of several autoimmune rheumatic

diseases.4 The clinical associations of these autoantibodies are given

in Table 1. These antibodies were traditionally detected using

immunodiffusion techniques, although ELISAs are now very widely

used. Enzyme digestion studies showed that these antigens were

sensitive to RNase and trypsin and are composed of ribonucleopro-

tein (RNP). The Sm (or Smith antigen; Sm, Ro and La antigens were

each named after the first two letters of the surname of the patients in

MEDICINE 38:2 70

whom the antibodies were first found) is resistant to RNase and

trypsin. The RNP antigenic determinants are found in association

with U1 RNA, whereas Sm determinants are present on U1, U2 and

U4-6 RNA. The other two important nuclear antigens are Ro and La.

Theseare identical to the independently identifiedantigens, Sjogren’s

syndrome A (SS-A) and Sjogren’s syndrome B (SS-B).

Antibodies against tRNA synthetases have been identified in

patients with polymyositis. The tRNA synthetases are cytoplasmic

enzymes that attach tRNA to its corresponding amino acid during

the assembly of polypeptides. Antibodies against several tRNA

synthetases have been described, including anti-Jo-1 (histidyl),

PL-7 (threonyl), PL-12 (alanyl), OJ (isoleucyl), EJ (glycyl) and KS

(asparginyl). Anti-Jo-1 (histidyl) is the most common, having been

found in up to 30% of patients with myositis. Patients with these

autoantibodies often share the same clinical features; the tRNA

synthetase syndrome e Raynaud’s phenomenon, fever, interstitial

lung disease, arthralgia and myositis. The tRNA synthetases are

located in the cytoplasm and hence are often not detected on

routine ANA testing, although a perinuclear staining pattern on IIF

ANA may be a clue to their presence.

Antitopoisomerase-1 antibodies (Scl-70) are associated with the

diffuse cutaneous variant of scleroderma and a high risk of pulmo-

nary disease. Topoisomerase-1 is a 100 kDa nucleolar enzyme

involved in uncoiling DNA prior to replication. Other enzymes tar-

geted in scleroderma (such as RNA polymerases I, II and III) are also

nucleolar and antibodies directed against these antigens are asso-

ciated with an increased risk of poor prognosis cardiac and renal

disease. Anticentromere antibodies (detected by IIF on HEp-2 cells)

are specific for the limited cutaneous variant of scleroderma and

react with the kinetochore of metaphase chromosomes.

Antiphospholipid (anticardiolipin) antibodies

Antibodies to phospholipids (particularly cardiolipin) are respon-

sible for the false-positive Wasserman reaction and lupus antico-

agulant seen inpatientswith theprimaryantiphospholipid antibody

syndrome (APS), and in some patients with SLE. The interaction of

phospholipid and a co-factor, b2-glycoprotein, is necessary for the


Disease associations of commonly detected autoantibodies

Autoantibody Disease Prevalence Disease

specificity

Associated clinical features

dsDNA SLE 70% High Lupus nephritis

Sm SLE 5% (Caucasian);

30e50%

(AfricaneCaribbean)

High Vasculitis, CNS lupus

Ro (SS-A) SLE 40% Low Photosensitivity, subacute cutaneous LE, congenital heart

block, neonatal LE

Sjogren’s syndrome 80% High Extra-glandular disease, vasculitis, lymphoma

La (SS-B) SLE 15% Low As for Ro

Sjogren’s syndrome 50% High As for Ro

U1RNP SLE 30% Low Raynaud’s phenomenon, swollen fingers, arthritis, myositis

(overlap syndrome e MCTD)

rRNP SLE 15% High CNS lupus (psychosis, depression)

PCNA (cyclin) SLE 5% High

Phospholipid SLE 35% High Thrombosis, foetal loss, thrombocytopenia

C1q SLE 40% Moderate Lupus nephritis

b2-glycoprotein APS 35% High Thrombosis, foetal loss, thrombocytopenia

Topoisomerase-1 Systemic sclerosis 30% High Diffuse cutaneous variant of scleroderma

Centromere Systemic sclerosis 30% Moderate Limited cutaneous variant of scleroderma, absence of lung disease

RNA-polymerases Systemic sclerosis 20% High Diffuse cutaneous variant of scleroderma, visceral involvement

PM-SC1 Systemic sclerosis 5% High Scleroderma/polymyositis overlap

Jo-1 Polymyositis 30% High Polymyositis with fibrosing alveolitis (anti-synthetase syndrome)

SRP Polymyositis 4% High Severe myositis

Mi-2 Polymyositis 10% High Dermatomyositis

CCP Rheumatoid

arthritis

80% High Rheumatoid arthritis

PR3 Vasculitis 90% High Wegener’s granulomatosis

MPO Vasculitis 50% Moderate Microscopic polyangiitis. ChurgeStrauss syndrome

GBM Anti-GBM disease >95% High Anti-GBM disease (Goodpasture’s syndrome)

CCP, cyclic citrullinated peptide; CNS, central nervous system; dsDNA, double-stranded DNA; GBM, glomerular basement membrane; LE, lupus erythematosus; MCTD,

mixed connective tissue disease; MPO, myeloperoxidase; PR3, proteinase 3; RNP, ribonucleoprotein; SLE, systemic lupus erythematosus; SRP, signal recognition

particle; SS, Sjogren’s syndrome.

Table 1


antigenic site to be available for binding of antibody. Anti-

phospholipid antibodies are associated with an increased risk of

vascular thrombosis, recurrent fetal loss, livedo reticularis and

thrombocytopenia. A positive IgG anticardiolipin test, present on

more than one occasion at least 6 weeks apart, is most specific for

APS but some patients are only IgM antibody positive.

Anti-b2-glycoprotein antibodies

Anti-b2-glycoprotein antibodies can also be detected by ELISA,

although this assay is less sensitive, but more specific than the

anticardiolipin antibody assay. The presence of anti-b2-glyco-

protein antibodies is now considered to be an important diag-

nostic criterion for APS.

Anti-C1q antibodies

C1q is the first component of the classical pathway of comple-

ment activation and hereditary deficiency is a risk factor for SLE.

MEDICINE 38:2 71

Antibodies against C1q are present in the serum of patients with

SLE (up to 47%) and are associated with renal involvement.

Monitoring anti-C1q may be useful to detect renal flares.8

Antineutrophil cytoplasm antibodies

Antineutrophil cytoplasm antibodies (ANCA) were first detected

in the sera of patients with systemic vasculitis in the early 1980s.

They are specific for granule proteins of neutrophils and mono-

cytes. Using IIF on ethanol-fixed human neutrophils, two major

staining patterns are observed: cytoplasmic (‘c-ANCA’) and

perinuclear (‘p-ANCA’). Two main antigenic targets have been

identified as proteinase 3 (PR3), which is associated with

cANCA, and myeloperoxidase (MPO), which is associated with

pANCA. The combination of cANCA and PR3 is highly

specific (>90%) for Wegener’s granulomatosis (WG), whilst

pANCAeMPO occurs in microscopic polyangiitis (MPA) and

ChurgeStrauss syndrome, but is less specific.9 A number of other

antigens (e.g. lactoferrin) are associated with a pANCA pattern



but have less clinical significance. International consensus

proposes that both IIF and ELISA for PR3 and MPO should be

performed on each sample.10 ANCA can be negative in patients

with limited WG, especially when limited to the upper respira-

tory tract. ANCA are also negative in other types of vasculitis,

such as giant cell arteritis, Takayasu arteritis, polyarteritis

nodosa and HenocheSchonlein purpura.

Anti-GBM antibodies

Antibodies directed against the glomerular basement membrane

(GBM) are highly specific for anti-GBM disease (Goodpasture’s

disease). These pathogenic antibodies are directed against an

epitope on the alpha-3 domain of type IV collagen, expression of

which is limited to the glomerular, retinal and alveolar basement

membranes. Anti-GBM antibodies (along with ANCA) should be

assayed in any patient presenting with a pulmonary-renal

syndrome or acute renal failure.

Practice points

C Anti-CCP antibodies are useful in making a diagnosis of RA

C Screening for ANA and ANCA should not be undertaken

routinely as the specificity is poor in this setting

C Rising concentrations of anti-dsDNA and ANCA indicate

possible relapse but should not be used as the sole reason for

a change in therapy in the absence of clinical features of

relapse

C Low concentrations of RF, ANA and ANCA are common in the

well elderly

Autoantibody measurement in the monitoring of disease activity

Serial detection of autoantibodies has a limited role in the

assessment of disease activity. Rising levels of dsDNA antibodies

in patients with SLE may correlate with increasing disease

activity. Similarly an increase in PR3-ANCA and MPO-ANCA

concentrations can occur in patients with active WG and MPA

respectively. However, in both situations the rise in antibody

concentration can precede clinical relapse by many months and

in some patients the concentrations fall immediately prior to the

relapse, suggesting tissue deposition. The detection of a rise in

concentration should serve to heighten suspicion and an increase

in the frequency of review, rather than a reflex increase in

immunosuppression.

When to request autoantibody tests

The detection of autoantibodies is a key part of the assessment of

the patient with suspected autoimmune rheumatic and renal

disease. ANA can be detected using HEp-2 cells in over 95% of

patients with untreated active SLE. Patients with autoimmune

disease can present with an ill-defined multi-system illness or be

non-specifically unwell. Such patients should be tested for the

presence of autoantibodies. However, screening for ANA and

ANCA should not be used indiscriminately. The probability of

a positive result is dependent on the prior probability of the

disease in question. Where this probability is low there is an

increased chance of a false-positive result. The positive predic-

tive value of ANA in SLE is around 11%. In addition, ANA are

not specific for autoimmune rheumatic disease and can occur in

other conditions such as organ-specific autoimmune disease

(primary autoimmune cholangitis, primary biliary cirrhosis),

chronic infection (subacute bacterial endocarditis), viral infec-

tion and lymphoproliferative disease. Detection of an autoanti-

body in an unexpected situation should prompt a careful clinical

review for features of the suggested disease. Equally, the absence

MEDICINE 38:2 72

of an autoantibody should not lead to the discarding of the

diagnosis if the clinical features are sufficiently clear-cut and

supported by other evidence such as histology.

About 15% of healthy adults and 8% of children have

detectable ANA, usually at low titre. The frequency of ANA in

normal people is higher in women and increases with age so that

around 25% of women over the age of 60 years have a positive

result. Most routine laboratories offer testing for the antibodies

described above. Specialist reference laboratories may also offer

a broader range of tests in line with their research interests. A

REFERENCES

1 Westwood OMR, Nelson PN, Hay FC. Rheumatoid factors what is new?

Rheumatology 2006; 45: 379e85.

2 Symmons DPM. Classification criteria for rheumatoid arthritis e time

to abandon rheumatoid factor? Rheumatology 2007; 46: 755e6.

3 Zendman AJW, van Venrooji WJ, Pruijn GJM. Use and significance

of anti-CCP antibodies in rheumatoid arthritis. Rheumatology 2006;

45: 20e5.

4 Damoiseaux JG, Cohen Tervaert JW. From ANA to ENA how to proceed?

Autoimmun Rev 2006; 5: 10e17.

5 Giles I, Isenberg DA. Antinuclear antibodies; an overview. In:

Wallace DJ, Hahn B, Hahn BH, Dubois EL, eds. Dubois’ lupus eryth-

ematosus. 5th edn. Baltimore: Lippincott, Williams and Wilkins,

2006: 432e41.

6 Bradwell AR, Stokes RP, Johnson GD. Atlas of Hep-2 cell patterns.

Birmingham: The Binding Site, 1995.

7 Isenberg DA, Manson JJ, Ehrenstein MA, Rahman A. Fifty years of anti-

dsDNA antibodies: are we approaching journey’s end? Rheumatology

2007; 46: 1052e6.

8 Trendelenburg M. Antibodies against C1q in patients with SLE.

Springer Semin Immunopathol 2005; 27: 276e85.

9 Hagen EC, Andrassy K, Csernok E, et al. The diagnostic value of

standardised assays for ANCA in idiopathic systemic vasculitis: results

of an international collaborative study. Kidney Int 1998; 53: 743e53.

10 Savige J, Gillis D, Benson E, Esnault V, Falk RJ, et al. International

consensus statement on testing and reporting anti-neutrophil cyto-

plasmic antibodies. Am J Clin Pathol 1999; 11: 507e13.


autoantibodies in the autoimmune rheumatic diseases

Documents